Wednesday, September 05, 2007
What is the latest theory of why humans lost their body hair?
This is the question asked in this month's issue of Scientific American. Mark Pagel, head of the evolutionary biology group at the University of Reading in England and editor of The Encyclopedia of Evolution gives three adaptationist explanations.
Now, here's the question of the day for all you adaptationists. Why didn't he mention neoteny? Do you think it's because he has carefully reviewed all the evidence and reaches the conclusion that there's more data to support running on the savannah?
Subscribe to:
Post Comments
(
Atom
)
19 comments :
Dawkins' TGD approaching one year on NYTimes bestseller list
Larry Moran: "Why didn't he mention neoteny?"
I don't know why Dr. Pagel chose not to mention it. However, a neotenic (nor any developmental-evolutionary) basis of a trait is not, by itself, an alternative explanation to an adaptationist hypothesis.
It is only an alternative if it is a nonadaptive side effect of another developmentally linked process. (And the frequency and intensity of that trait is itself likely under positive selection.) For example, Gould's global heterochrony hypothesis. In Gould's model, neotenization itself is adaptive because it enables cognitive flexibility but the specific neotenic manifestation of hairlessness (and a host of other traits) is a nonadaptive epiphenomenon of that adaptation. (At least from what I recall of Ontogeny & Phylogeny. My copy is in storage.)
Invoking any form of heterochrony, homeosis, etc. is not, by itself, an alternative to adaptation. In fact, these may be developmental mechanisms of adaptation. (See Tinbergen's Four Questions for why this is so.) In Gould's particular model it is an alternative (as far as hairlessness is concerned).
Of course, such traits may not be adaptive, but their developmental bases are not by themselves determinative of whether or not they are adaptive.
somebody hand me that ten-foot pole...
It's only fair to note that human neoteny is merely a fairly popular (but controversial) hypothesis, not an established fact. Perhaps Pagel is a nonbeliever.
Steve LaBonne notes,
It's only fair to note that human neoteny is merely a fairly popular (but controversial) hypothesis, not an established fact.
He mentions three adaptationist hypotheses that are not established facts.
Perhaps Pagel is a nonbeliever.
He indicates his preference for one of the adaptationist hypotheses but still mentions the others. He seems to be trying to explain to the questioner the current scientific ideas about the origin of hairlessness. So why didn't he mention a non-adaptationist hypothesis in the interests of fairness?
I'm pretty sure you know what I'm getting at. Pagel is an adaptationist and adaptationists have a very myopic view of evolution. Non-adaptationist explanations just aren't on their radar. They don't merit consideration.
I didn't read the article, I just skimmed it to see what his three hypotheses are:
1. Aquatic ape
- thoroughly debunked by now, I thought
2. Body temperature regulation on open grasslands
- This gets at effect vs. function
3. Fur-dwelling parasite avoidance
- I've not seen this one before
Rather than discuss those three hypotheses, I'd like to add another, probably blatantly adaptationist, hypothesis not relating to natural selection.
4. Sexual selection
This is based on much of what Desmond Morris talks about in his 1969 book The Naked Ape. I'm not going to go into detail here, except to note that I find it curious that sexual selection is so neglected among the adaptationists. Also, to point out that discussions of various competing hypotheses to explain the only great ape lacking extensive body hair are not new.
Having said all that, I'd like to assert that I do not fit Dr. Moran's definition of an adaptationist - I think the four primary "forces" of evolution all contribute to the evolution of populations, and are central to modern evolutionary biology theory. I might or might not differ from Dr. Moran in my opinions of the relative importances of those four forces.
Refresher:
Selection (natural and sexual)
Drift
Mutation
Migration
I'd also like to further the point by anonymous, above, that neotony isn't necessarily a non-adaptationist answer. Am I incorrect in thinking that neotony is a phenotypic trait, and as such is subject evolutionary change via the above mechanisms? The pleiotropy hypothesis advanced by anonymous is also quite good, I think.
Non-adaptationist explanations just aren't on their radar.
Again, for all you know he has long since considered and rejected the hypothesis that humans are neotenous. So in this case the verdict is "not proven", as they say in Scotland.
mostly for the benefit of those who missed out on the same exact argument on the other thread, allow me to represent the pluralist-like-everybody-else but leaning-strongly-to-the-adaptionist-
camp-when-discussing-phenotypes point of view:
"Neoteny," as pointed out above, is about the mechanism of development for hairlessness. By itself it is insufficient to explain the evolution of hailessness. Rather, for neoteny to be involved, one of the following four combinations of conditions must hold:
1) neoteny + (near) neutrality + drift
2) neoteny + a severe bottleneck or founder event + insufficient time (for whatever reason) for neotenic hairlessness to be reversed by directional selection (i.e. hairlessness has negative, not positive or neutral survival value)
3) neoteny + natural section for hairlessness
4) sexual selection for hairlessness in women, with mail hairlessness analogous to male nipples (developmental constraint limiting sexual dimorphism, or whatever).
Note that these explanations, except for 3 & 4, are mutually exclusive. Apparently, Dr. Moran favors #1, without evidence. Here, very briefly, is evidence for #3 (and therefore against # 1):
I doubt the most knee-jerk pluralist would argue with the statement that in mammals generally, fur functions as thermoregulatory insulation. Empirically, the thickness of the pelage has a large effect on the energy budget of mammals, especially those with body size in the middle size range, including human-sized. Conventional energy-allocation arguments, backed up with good data, suggest that hair loss is extremely unlikely to be a neutral trait.
Human bodies are very poorly designed for retaining heat. Therefore,in even moderately cool environments, naked humans have to spend large amounts of energy on thermoregulation. To argue for fitness-neutrality requires either a climate near the human thermoneutral zone (depending on water availability on the high end, about 30-40C) nearly all the time, or the invention of clothing and/or fire before or simultaneously with the appearance of hairlessness.
On the other hand, the human body is very well designed for losing heat from the entire body surface in an environmental temperature higher than dry skin temperature. This ability is directly attributable to hairlessness, especially in combination with the dense distribution of apocrine sweat glands on the entire body surface. Both traits differ markedly from extant chimpanzees.
Evidence against #2: all of the above plus the known human paleoclimate.
Evidence for #4: The geographic differences among human populations in hairiness, plus the sexual dimorphism in hairiness (therefore hairlessness) within all populations.
Evidence against #4: Intersexual selection of female traits by male choice is very unusual. This is an area of current research in behavioral ecology.
Evidence for #1 or #2: I am honestly unaware of any.
So that's as clear as I can state the reasons I am an "adaptationist" in the case of human hairlessness. But, I hasten to repeat, we are all pluralists, even if we argue about specific cases.
From the comments:
"Having said all that, I'd like to assert that I do not fit Dr. Moran's definition of an adaptationist...."
and
"But, I hasten to repeat, we are all pluralists, ...."
Wrong, wrong, wrong. You chaps have no idea how wrong you are, for it is written:
At this point, some evolutionists will protest that we are caricaturing their view of adaptation. After all, do they not admit genetic drift, allometry, and a variety of reasons for nonadaptive evolution? They do, to be sure, but we make a different point. In natural history, all possible things happen sometimes; you generally do not support your favored phenomenon by declaring rivals impossible in theory. Rather, you acknowledge the rival but circumscribe its domain of action so narrowly that it cannot have any importance in the affairs of nature. Then, you often congratulate yourself for being such an undogmatic and ecumenical chap.
Pete Dunkelberg
Steve DiMilo says,
"Neoteny," as pointed out above, is about the mechanism of development for hairlessness. By itself it is insufficient to explain the evolution of hailessness. Rather, for neoteny to be involved, one of the following four combinations of conditions must hold:
I thought the idea was that there was selection for larger brains and this was satisfied by shifting the growth and development of the fetus so that there was a longer period of growth after birth.
Hairlessness then became an unintended consequence (epiphenomenon) of selection for neoteny.
Was that one of your four choices?
Hairlessness then became an unintended consequence (epiphenomenon) of selection for neoteny. Was that one of your four choices?
Hmmmm, no, I guess not...good point. So with regard to hairlessness itself, this situation would either be a subset of #1 (hairlessness itself is neutral--I have tried to explain why I doubt that is true) OR you have to postulate some ineluctable pleiotropic or genetic linkage of body-hair growth to brain growth or to development in general. I guess "in natural history, all possible things happen sometimes" but in this case at least such a link and its ineluctability are of course pure speculation.
"you generally do not support your favored phenomenon by declaring rivals impossible in theory. Rather, you acknowledge the rival but circumscribe its domain of action so narrowly that it cannot have any importance in the affairs of nature. Then, you often congratulate yourself for being such an undogmatic and ecumenical chap.
Ouch! Gould and Lewontin I presume. I can only say that it's not me who restricts drift to neutral traits (in the absence of linkage or pleiotropy); it's the definition of drift itself.
And I'm happy to cop to the label "adaptationist" if it makes you feel good inside.
mostly for the benefit of those who missed out on the same exact argument on the other thread, allow me to represent
Excellent! This was opening a large window into biology thinking. In fact, the whole subject area Larry has chosen to discuss is a gold mine. (For us who is too busy or lazy to mosey down to the book store and stock up on Dawkins, Gould, et cetera.)
Here, very briefly, is evidence
Okay, here I run up against a perceived difference with my area, physics. "Evidence", at least for me, is testable evidence. I even had to check a dictionary - I see that "things helpful in forming a conclusion" or "something indicative" is indeed the broader use. So this is circumstantial evidence.
But since it isn't testable it is merely and IMO more rightly support for the original hypothesis. Formally it goes into reformulating a new and stronger hypothesis, roughly "Since our native climate was likely hot, we lost hair due to this specific selective pressure".
And then we try to make a prediction that we can test. [Obviously IANAB, but suppose that we should be able to correlate closely with increased density of those apocrine sweat glands by gene dating if we can find specific genes. While shifting growth period as null hypothesis wouldn't correlate.]
I'm a bit hesitant to mention this because it could unintended be taken as personal criticism. But I really want to note that I don't perceive biology as bad theory builders.
On the contrary, it kicks ass! For example, I have only had reason to read short pieces of Darwin due to creationist discussions, but I was stricken by his ability to establish his data and isolate what he could prove.
[Specifically, it was that bit about specifying "variation" instead of "random mutations" (which of course didn't exist in the genomic form at his time) or some other artificial restriction in his proposed mechanism. One can see the best scientists or mathematicians taking much care to get the basic definitions just right. Einstein and Feynman are other accessible examples of this - mathematicians are more a rumor because, damn, modern math... :-P]
Sven DiMilo says,
Ouch! Gould and Lewontin I presume. I can only say that it's not me who restricts drift to neutral traits (in the absence of linkage or pleiotropy); it's the definition of drift itself.
Nope. Random genetic drift refers to changes in the frequency of alleles that are independent of selection. It applies to deleterious alleles that can be fixed by drift, to neutral alleles that can be fixed or lost, and to beneficial alleles that are lost. You are aware, I hope, that most beneficial alleles do not reach fixation?
If the selection coefficient is 0.1 then the probability of fixation is only 20%. The other 80% is random genetic drift at work.
I happened on what seems to me to be another informative post by paleoanthropologist John Hawks. He discusses what biologists think is important to describe, problems with null hypotheses, and problems with "just-so stories":
For example, Milford Wolpoff and I (2001) showed that the rate of evolution of brain size in Late Pleistocene Neandertals was probably directionally selected. Yet, we could not reject the hypothesis of drift, under the condition that the Neandertal population was as small as 1200 effective individuals. Keep in mind that brain size evolution in humans is one of the most rapid evolutionary trends known for any large mammal! The hypothesis of drift is very hard to refute statistically with paleontological data, particularly in the absence of reliable data about demography.
Treating selection on an equal footing is the only alternative. The reason why we formulate adaptive hypotheses about phenotypes is to propose connections between traits. Only by including multiple traits in our test can we avoid the problem that our statistics are insufficient to differentiation selection from drift based on the rate of change alone. The reason why evolutionary biologists ought to deride "just-so stories" is not that they are "silly", but instead because adaptive hypotheses of this kind generally do not propose testable connections between characters. [My emphasis.]
He disses other tests of selection as well, but also discusses lack of tests for drift (at least in his subfield):
Even though I have seen many papers where a slow rate of change was attributed to drift, I have never seen a single one test for a rate too low to be drift (and therefore indicative of stabilizing selection). Given these problems, the "burden of proof" idea is important to keep in mind, because it is so rare that the hypothesis of drift is tested in more than one way. [His emphasis.]
IANAB, but on the need for testing, testing, testing, I can't help but agree.
You are aware, I hope, that most beneficial alleles do not reach fixation?
I am; everything you say is of course correct and I stand corrected.
If the selection coefficient is 0.1 then the probability of fixation is only 20%. The other 80% is random genetic drift at work.
Uncle! Like Jake Blues, I have Seen the Light. I am hereby coming out as a born-again, card-carrying Pluralist.
so...where do I pick up my card?
Torbjörn, your link leads back to this post. Here's the correct link:
http://johnhawks.net/weblog/topics/evolution/selection/pigliucci_lynch_review_2007.html
From your saying that he "disses tests of selection" one might get the impression that he is skeptical of selection. On the contrary, he is saying that selection is a lot more important than currently available tests of selection let on. (Sorry if I'm telling you something you already know, just a small clarification to the bits you quoted.)
windy:
Thanks for the corrections!
Yes, I could have expanded on Hawks post. But I felt I was longwinded enough. :-P
Dears,
Pagel hypothesis explains both human, mole rats (and perhaps some kinds of bats) loss of fur.
All are very social animals that could suffer from ectoparasites plagues.
Also, Humans are the only primate that have lices (because, contrary to other primates, we are more sedentary, enabling the completion of a full reprodutive cycle of lices).
The neoteny hypothesis only explain the human case. Point for the ectoparasite case!
Post a Comment